Hydrogels for sustained delivery of biologics to the …kinampark.com/KPTopics/files/Hydrogels, General/Chang...REVIEWS Drug Discovery Today Volume 24,Number 8 August 2019 Teaser significant

Reviews�FO

UNDATIO

NREV

IEW

REVIEWS Drug Discovery Today �Volume 24, Number 8 �August 2019

Teaser Hydrogels providing sustained delivery of biologics to the back of the eye can make asignificant impact in the treatment of many ocular diseases. This article reviews the

properties, administration approaches, and development challenges for an ocular hydrogeldelivery system.

Hydrogels for sustained delivery ofbiologics to the back of the eyeDebby Chang1, Kinam Park2 and Amin Famili3

1Drug Delivery, Genentech, Inc., South San Francisco, USA2Departments of Biomedical Engineering and Pharmaceutics, Purdue University, West Lafayette, USA3 Small Molecule Analytical Chemistry, Genentech, Inc., South San Francisco, USA

Hydrogels are water-laden polymer networks that have been used for

myriad biological applications. By controlling the chemistry through

which a hydrogel is constructed, a wide range of chemical and physical

properties can be accessed, making them an attractive class of

biomaterials. In this review, we cover the application of hydrogels for

sustained delivery of biologics to the back of the eye. In adapting hydrogels

to this purpose, success is dependent on careful consideration of material

properties, route of administration, means of injection, and control of

drug efflux, all of which are addressed. We also provide a perspective on

clinical and chemistry, manufacturing and controls (CMC) considerations

that are integral to the development of an ocular hydrogel delivery system.

IntroductionOcular drug delivery of biologics to the posterior segment of the eye is an important and rapidly

developing field due to the growing need for treatments for ocular diseases such as age-related

macular degeneration (AMD), retinal vascular diseases, posterior uveitis, and glaucomatous optic

neuropathies [1]. Durable and effective drug delivery to the back of the eye remains a significant

challenge. Delivery of biologics to posterior ocular targets is made especially challenging due to

anatomical and physiological barriers, resulting in poor bioavailability at posterior tissues via

anterior or systemic delivery routes [2]. The relatively small volume of the vitreous humor also

dictates that delivery of drugs into the space are limited in volume to less than or equal to 150 mL

[3], and that drugs clear relatively rapidly from this space into systemic circulation [4]. Innovative

drug delivery approaches from nanoparticles, liposomes, micelles, hydrogels, dendrimers, micro-

particles, nanotubes, and implants have been evaluated. Few have made it to late stage clinical

trials, with some recent exceptions such as the nondegradable implant reservoir, Port Delivery

System [2]. Finding biodegradable drug delivery solutions that provide durable exposure to

posterior tissues in a tractable delivery format remains a major unmet need.

Biologic drugs, whether isolated from natural sources or manufactured by recombinant DNA

technology (i.e., biotechnology), are sensitive to environmental conditions including pH,

Debby Pei-Shan Chang

is a Technical

Development Scientist in

the Drug Delivery

department at Genentech.

She has been involved in

the polymer-based delivery

system for large molecules,

peptides, and small

molecules. She received her B.S. degree in

Bioengineering from U.C. Berkeley, Ph.D. in

Mechanical Engineering & Materials Science (MEMS)

from Duke University, and completed her

postdoctoral fellowship in Physical Chemistry

department at Lund University, Sweden.

Kinam Park is the

Showalter Distinguished

Professor of Biomedical

Engineering and Professor

of Pharmaceutics at Purdue

University. He has studied

drug delivery systems for

three decades with the

research focus on the use

of polymers for controlled release formulations. His

current research focus is PLGA-based injectable, long-

acting formulations. He is the founder of Akina, Inc.

specializing in specialty polymers used in drug delivery

systems and biomedical devices. He serves as the

Editor-in-Chief of the Journal of Controlled Release.

Amin Famili is a Scientist

in the Small Molecule

Analytical Chemistry

department at Genentech.

He received his Ph.D. from

the University of Colorado

in 2014 and completed his

postdoctoral fellowship in

the Drug Delivery

department at Genentech in 2016. His work has

focused on the design, characterization and

development of polymer-based drug delivery systems,

with a specific focus on ocular delivery of biologics and

peptides.

Corresponding author: Famili, A. ([email protected])

1470 www.drugdiscoverytoday.com1359-6446/ã 2019 Elsevier Ltd. All rights reserved.

https://doi.org/10.1016/j.drudis.2019.05.037

http://crossmark.crossref.org/dialog/?doi=10.1016/j.drudis.2019.05.037&domain=pdf

mailto:[email protected]

https://doi.org/10.1016/j.drudis.2019.05.037

Drug Discovery Today �Volume 24, Number 8 �August 2019 REVIEWS

Review

s� F

OUNDATION

REV

IEW

temperature, salt concentration, shear forces, and exposure to

water/solid/air interfaces, further complicating the afore-men-

tioned design constraints for posterior ocular drug delivery [5].

Because of this, hydrogels can be seen as a good choice for

encapsulation and release of biologics [5]. Hydrogel properties

can be tuned over a relatively wide range and can be designed

to provide a close mimic for biological matrices, exhibiting many

similar properties including high hydrophilicity, high hydration,

and closely matched mechanical properties [6]. These character-

istics may also serve to preserve the physicochemical state of

biologic drugs over extended times, especially when compared

to more rigid, hydrophobic polymer matrices that can be used for

drug encapsulation and release (e.g. well-reported degradation of

peptides and proteins in solid matrices based on biodegradable

poly(lactide-co-glycolide) (PLGA)) [7].

Various hydrogels have been used in a large number of drug

delivery systems, in particular for oral drug delivery [8,9]. Hun-

dreds of oral sustained release drug delivery systems exist for

clinical applications. On the other hand, the use of hydrogels

for non-oral routes of delivery have not been as extensive, and in

fact, have been very limited. In the realm of ocular drug delivery,

despite a long record of literature reports of hydrogels designed for

this purpose, there remains no marketed hydrogel-based ocular

drug delivery products. This absence is due primarily to the

challenge of meeting a strict set of performance requirements

including safety, tolerability, manufacturability, degradability

and ease of administration, while maintaining activity of the

encapsulated drug and achieving a clinically-relevant drug release

profile.

Despite the absence of marketed products, the promise of

hydrogels as ocular drug delivery systems remains. Hydrogels

are generally considered to be highly biocompatible due to the

excess of water present within the system, which is thought to

mimic biological matrices [10]. Hydrogels can also be designed to

provide short- or long-term release of drugs including peptides and

proteins [11–16], whether by physical or chemical control. Mod-

ern-day hydrogels, including ‘‘smart” or environment-responsive

hydrogels [17], have properties that were not available decades

ago, and hydrogels with improved properties have made it possible

to consider ocular drug delivery for durations of weeks or months

while overcoming challenges of limited modes of administration.

In this review, we describe hydrogel characteristics and physio-

chemical properties for delivery of biologics to the back of the eye.

The physiological barriers of the eye and the applications of

hydrogel via different routes of administration are discussed.

Moreover, different classes of hydrogels, from stimulus responsive

to injectable in situ forming gels, are reviewed. In addition, clinical

and manufacturing challenges anticipated during the develop-

ment of a hydrogel ocular delivery system are discussed.

Hydrogel characteristicsA hydrogel is a three-dimensional network of hydrophilic poly-

mers that absorbs water and swells but exists as a solid material.

Hydrophilic polymers interact favorably with water molecules

through backbone or pendant hydrophilic chemical structures,

generally resulting in a high water solubility. In a hydrogel,

hydrophilic polymer chains are made to form a network which

retains the favorable water interactions but resists dissolution

through chemical or physical crosslinks. A crosslinked three-di-

mensional polymer network swells in water for the same reason

that an analogous linear polymer mixes with water spontaneously

to form an ordinary polymer solution; the swollen hydrogel is in

fact an elastic solution rather than a viscous one [18]. Hydrophilic

polymers that are commonly used to form hydrogels include

synthetic polymers, such as poly(hydroxyethyl methacrylate)

(PHEMA), poly(ethylene oxide) (PEO, also called poly(ethylene

glycol) (PEG)), polyvinylpyrrolidone (PVP), poly(acrylic acid)

(PAA), polyacrylamide (PAM), and poly(vinyl alcohol) (PVA),

and natural polymers include agar, gelatin, fibrin, hyaluronic acid

(HA), alginic acid, carboxymethylcellulose (CMC), and hydroxy-

propyl methylcellulose (HPMC).

A hydrogel network can be formed by either covalent or non-

covalent bonds among polymer chains to form chemical or physi-

cal hydrogels, respectively [19]. Examples of non-covalent bonds

that can be engaged include hydrogen bonding, hydrophobic

interactions, and physical entanglements. A hydrogel can be a

homogeneous network of the same polymer, or it can a mixture of

two or more different polymers. If each of the two different types

of polymers makes its own network through covalent bonds, the

overall structure is called an interpenetrating network. If one of

the two networks is not covalently crosslinked, the structure is

known as a semi-interpenetrating network.

History of hydrogelsAccording to SciFinder1, the first articles that used the term

‘‘hydrogel” were published in the mid-1890s to describe gelatinous

ferric hydroxide [20] of silica gel [21]. Hydrogels made of hydro-

philic polymers were first described in 1900 using the alcohol-

gelatin-water ternary system [22]. The concept of hydrogels that

we deal with nowadays, however, began in 1960 when Wichterlie

and Lim published their paper on hydrophilic gels for biological

use [23]. Hydrogels in their early stage of development possessed

only one function: absorbing water and swelling. Recent advances

in polymer chemistry have produced numerous new hydrogels

that possess additional functions, such as the ability to respond to

changes in environmental factors. These ‘‘smart” hydrogels swell,

change shape, undergo sol-gel phase transition, or degrade in

response to various environmental stimuli [24,25]. These proper-

ties have been used to develop stimulus-responsive drug delivery

systems, as will be described in a later section.

Physicochemical properties of hydrogelsPhysical properties of hydrogelsWater absorption and subsequent swelling in the presence of

abundant water is the most fundamental property of hydrogels.

The extent of water absorption, and thus, swelling of a hydrogel,

varies significantly between hydrogel chemistries. The extent of

swelling is usually measured by the swelling ratio (Q) which is

defined by the weight (or volume) of a swollen hydrogel (Qs)

divided by the weight (or volume) of a dried gel (Qd). The Q value

of hydrogels can vary from slightly higher than 1 to more than

100. There is no predefined extent of swelling that makes a

material a hydrogel. The extent of swelling depends on many

factors, including the nature of hydrophilic groups and degree of

crosslinking. As the crosslinking density increases, the swelling

ratio decreases. Thus, whether a material can be considered a

www.drugdiscoverytoday.com 1471


TABLE 1

Examples of external factors affecting hydrogel properties

Chemical factors Physical factors Biological factors

pH Temperature AntibodiesIon type Light EnzymesIonic strength Electrical field GlucoseOrganic solvent Magnetic field Ligands

Ultrasound

Reviews�FO

UNDATIO

NREV

IEW

hydrogel is not based on the extent of swelling or the amount of

absorbed water. Rather, it is based on the nature of polymer chains.

If polymer molecules in a hydrogel can dissolve in water in the

absence of crosslinking, the crosslinked network can be called a

hydrogel. In the presence of a limited amount of water not suffi-

cient to dissolve all the polymers, even non-crosslinked polymers

in a partially swollen state can be called hydrogels.

Because a hydrogel is a crosslinked network, it behaves like a

solid material, and thus, it does not flow. An exception to this can

occur if the crosslinking density of a hydrogel is so low that the

solid characteristics cannot be maintained. At very low crosslink-

ing density, polymer chains behave more like a viscous solution

than a gel. Even in the absence of any crosslinking, hydrophilic

polymers can form a transient hydrogel. For example, a matrix of

compressed hydrophilic polymers, such as poly(acrylic acid) or

hydroxypropyl methylcellulose (HPMC), can absorb a small quan-

tity of water to form a hydrogel network. However, this network is

only transient; as more water is absorbed, polymer chains disen-

tangle and the network dissociates. This transient crosslinking

behavior can be either an advantage or a disadvantage depending

on the intended application.

Sensitivity to external stimuliHydrogels can be designed to undergo a pre-defined change in

response to external stimuli [17]. As illustrated in Fig. 1, hydro-

gels can swell or deswell, degrade, change shape and undergo

phase transitions in response to environmental or external sti-

muli. Hydrogel swelling and deswelling can be dictated by

changes in hydrophobic interactions as a function of tempera-

ture. Hydrogel degradation can occur through cleavage of bonds

within the hydrogel network either through enzymatic or chem-

ical means. Hydrogels can change their shape in response to

physiological conditions, e.g., if two hydrogels with different

swelling properties are layered. Finally, hydrogels can transition

between solution phase and gel phase based on physical cross-

linking among polymer chains. For example, gelatin can act as a

thermosensitive hydrogel, since it dissolves at room temperature

Deswelling

Sol-gelPhase transition

FIGURE 1

Hydrogels can be designed to respond to various environmental factors. A hydrogechange. These processes are generally repeatable with the exception of degrada

1472 www.drugdiscoverytoday.com

but forms a gel at lower temperature. The inverse case, referred to

as inverse thermosensitive hydrogels, are soluble at lower tem-

peratures and transition to a gel with increasing temperature.

These polymers typically possess hydrophobic moieties, the

hydrophobic character of which increases with temperature. At

a certain temperature, these interactions will result in

formation of a physical network through hydrophobic chain

entanglements [26].

There are various external factors that can be used to alter the

properties of smart hydrogels (Table 1). The external factors can be

grouped into three categories: chemical, physical, and biological.

Various environmentally sensitive polymers have been used for

ocular drug delivery for enhanced ocular retention of the admin-

istered drug and sustained release [27]. Of the factors listed in Table

1, temperature may be the most relevant for ocular drug delivery. A

polymer solution at room temperature can be administered to the

eye to form a hydrogel at body temperature. Commonly used

temperature-responsive polymers are tri-block copolymers of PEO

and poly(propylene glycol) (PPG), which are commercially avail-

able under Poloxamers (manufactured by ICI) and Pluronics (man-

ufactured by BASF), N-isopropylacrylamide copolymers, and

copolymers of PEO and poly(lactic-co-glycolic acid) (PLGA) [28].

Routes of hydrogel administrationOcular drug delivery is a particular challenge due to the unique

physiological and anatomical barriers of the eye [2]. Various forms

of ocular drug delivery systems have been developed for clinical

use and pre-clinical investigation and can be divided based on the

Degradation

Shapechange

Drug Discovery Today

l can undergo deswelling, degradation, gel-to-sol phase transition, and shapetion.


Review

s� F

OUNDATION

REV

IEW

routes of administration into systemic delivery, extraocular deliv-

ery, and intraocular delivery [29]. Extraocular delivery occurs

either by topical delivery or subconjunctival delivery. On the other

hand, there are many methods for intraocular delivery, including

intrastromal, intracameral, intrascleral, intravitreal, suprachoroi-

dal, and subretinal delivery [29], as shown in Fig. 2. The relevant

ocular barriers depend on the route of administration. The corneal

epithelium is the primary tissue barrier restricting drug penetra-

tion after topical administration, the conjunctival epithelium and

associated tight junctions restrict subconjunctival administration,

and the internal limiting membrane (ILM) and retinal pigment

epithelium (RPE) are the limiting barriers for intravitreal adminis-

tration [30]. The efficacy of a specific drug delivery system will

depend in part on the route of administration, the pharmacoki-

netic parameters (such as bioavailability and clearance) at the

target site, and the potency and release rate of the drug [31].

Protein permeability through different ocular tissues have been

reviewed [32] and this can greatly affect the uptake and clearance

of the biotherapeutic. The use of hydrogels as drug delivery

vehicles can enhance the bioavailability by increasing the contact

time or serve as a sustained release depot for the drug. The uses of

hydrogels through different routes of delivery are discussed below

(Table 2).

Topical delivery from hydrogelsTopical delivery, e.g. via ophthalmic drops or contact lenses, is the

most non-invasive route of delivery. However, due to limited

penetration to posterior tissues, topical delivery is typically used

for external, corneal, and anterior segment diseases, and only a few

studies have suggested efficacy for posterior segment diseases [29].

One of these studies showed intraocular penetration of a single

chain antibody with a molecular weight (MW) of 26 kDa after

topical application in a rabbit model [33]. Intraocular penetration

to the posterior and anterior segment of the eye was found to occur

with a frequent dosing regimen with and without a penetration

enhancer. Topical delivery of protein therapeutics to the posterior

segment is limited to peptides or proteins with low MW [34]. It

requires highly concentrated formulations because of the low

ScleraC

Conjunctiva

Topical

Subconjunctival

unctiva

FIGURE 2

Hydrogels have been delivered via various routes of administration to access differethe hydrogel (e.g. monolithic depot vs. micro/nano-gel) and the therapeutic nee

bioavailability (<5%) arising from poor drug penetration and loss

from tear lacrimation [2].

For topical delivery, hydrogels can be used to increase the

contact time of the drug formulation with the cornea. Examples

include hydrogels in the form of contact lenses or viscous

solutions. Several studies have evaluated the delivery of proteins

to the posterior segment of the eye through hydrogel contact

lenses [35–38]. Rabbits fitted with drug loaded hydrogel contact

lenses showed detectable drug levels in the posterior segment of

the eye for both small molecule drugs and larger biological

molecules such as ranibizumab [37]. A clinical study using

the Boston Ocular Surface Prosthesis (BOSP) lens, a large diame-

ter, rigid, gas permeable lens, to deliver bevacizumab

showed improved visual acuity [35]. The BOSP contact lens

improved bioavailability by prolonging the drug exposure time

to the eye.

Intravitreal delivery from hydrogelsIntravitreal (IVT) injection is performed by direct injection of drug

into the vitreous chamber of the eye. IVT has been the preferred

route of administration for posterior targets because it can achieve

high drug concentrations in the vitreous and potentially high

bioavailability to posterior tissues such as the retina. The injection

procedure is typically done in a clinical setting with a 27 or 30 G

needle. The injection volume in humans is typically equal to or

less than 100 mL to limit the transient elevation in intraocular

pressure that results from an intravitreal injection [3].

IVT injection of biologics still requires repeated injections,

which carry the risk of complications such as endophthalmitis,

retinal detachment, iritis/uveitis, intraocular hemorrhage, ocular

hypertension, cataract, and hypotony [39]. Novel long-term con-

trolled release systems, such as drug-encapsulating hydrogels, may

reduce the injection frequency and improve patient compliance.

IVT injection of a hydrogel for long-term delivery of proteins

requires the hydrogel system to be injectable, biodegradable, and

biocompatible. Most importantly, the hydrogel system must dem-

onstrate sustained delivery of protein therapeutics for up to several

months. The hydrogel delivery system can be an in situ-gelling

horoid

Retina

Suprachoroidal

Intravitreal


nt sites of action. The route of administration is dictated both by the form ofds of the drug being delivered (e.g. site of action in retina vs. choroid).



TABLE 2

Examples of hydrogel delivery systems designed for sustained release of biologics

Hydrogel System Drug Releaseduration

Release Mechanism Remarks References

Alginate microsphereswithin a collagen hydrogel

BSA 11 days Diffusion Alginate microsphereencapsulated within a collagenhydrogel for sustained deliveryof BSA

Liu, 2008 [84]

2-hydroxyethyl methacrylateand 2-aminoethylmethacrylate p(HEMA-co-AEMA)

BSA 12 days Diffusion and swelling Nanocomposite contact lensmade of gelatin nanoparticlegrafted onto p(HEMA-co-AEMA)to encapsulate hydrophilicprotein drug

Zhang, 2013 [85]

Tetra-PEG with ß-eliminativelinkers

Exenatide, peptide a few hours toover a year

Self-cleaving linker ß-eliminative linkers to tetherdrugs and crosslink PEGhydrogel for sustained releaseand controlled hydrogeldegradation

Ashley, 2013 [42]

pNIPAAm-dextran BSA 15 days Diffusion throughthermoresponsivehydrogel

Thermoresponsive andbiodegradable hydrogel foraqueous encapsulation andrelease of hydrophilic drug

Huang, 2005 [86]

4-arm PEG Bevacizumab 14 days Diffusion and hydrogeldegradation

In situ crosslinked PEG hydrogelfrom 4-arm PEG-Mal and 4-armPEG-SH

Yu, 2014 [68]

Polycaprolactonedimethacrylate (PCM) andhydroxyethyl methacrylate(HEMA)

Bevacizumab 4 months Diffusion and hydrogeldegradation

Light activated, in situ formingPCM and HEMA based hydrogelfor sustained suprachoroidaldelivery of bevacizumab

Tyagi, 2013 [48]

Crosslinked alginate andchitosan

Lysozyme Bevacizumab, 18 days3 days

Diffusion and erosion Injectable in situ crosslinkedpolysaccharide hydrogel

Xu, 2013 [41]

pNIPAAm crosslinked withPEG-DA

BSA, IgG 21 days Diffusion throughthermoresponsivehydrogel

Crosslinked thermoresponsivehydrogel

Kang Derwent, 2008[60]

PEG-heparin hydrogel Heparin 49 days Retro-Michael reaction Hydrogel made with reversiblemaleimide-thiol linkage that issensitive to reducingenvironments

Baldwin & Kiick 2013[87]

Triblock copolymer (PEO)z-PCL-(PEO)z

Bevacizumab 20 days Diffusion and hydrogeldegradation

Thermal responsivebiodegradable triblockcopolymer with reversible sol-gel transition for drugencapsulation and sustainedrelease

Wang, 2012 [61]

Hydrogel contact lenses Ranibizumab 11 days Diffusion Hydrogel contact lenses fordelivery of ranibizumab

Schultz, 2011 [37]

PEG-poly-(serinolhexamethyleneurethane) (ESHU)

Bevacizumab 9 weeks Diffusion Thermal responsive ESHU gel forsustained release ofbevacizumab

Rauck, 2014 [40]

PNIPAAm crosslinked withAcryl-PLLA-PEG-PLLA-Acryl

IgG, Bevacizumab,Ranibizumab

10 days Diffusion throughthermo-responsivehydrogel

Glutathione-modulateddegradation of and release frompNIPAAm-based thermalresponsive hydrogels

Drapala, 2014 [56]

Enzymatically responsivePEG hydrogel

Peptide 10 days Enzymatic (MMP)cleavage

PEG hydrogel withenzymatically degradablesequence for delivery of peptidedrug

Van Hove, 2014 [43]

Pentablock copolymer IgG-Fab 80 days Diffusion and hydrogeldegradation

Pentablock copolymer (PCL-PLA-PEG-PLA-PCL) nanoparticlesuspended in thermalresponsive copolymer gel

Agrahari, 2016 [88]


Reviews�FO

UNDATIO

NREV

IEW


TABLE 2 (Continued )

Hydrogel System Drug Releaseduration

Release Mechanism Remarks References

Crosslinked hyaluronic acid/dextran

Bevacizumab 6 months Diffusion and hydrogeldegradation

In situ hydrogel formed bycrosslinking betweenvinylsulfone functionalizedhyaluronic acid (HA-VS) andthiolated dextran(Dex-SH)

Yu, 2015 [44]

PLGA microspheressuspended within PNIPAAm-PEG-DA hydrogel

RanibizumabAflibercept

200 days Diffusion and hydrogeldegradation

Injectable, thermal responsivemicrosphere-hydrogelcombined system

Osswald, 2016, [89]2017 [90]

PLGA microsphere in PEG-PLLA-DA/NIPAAm

RanibizumabAflibercept

180 days Diffusion and hydrogeldegradation

Physical and mechanicalcharacterization ofbiodegradable microsphere-hydrogel system

Liu, 2018 [77], 2019[78]

pNIPAAM: poly(N-isopropylacrylamide); PEG: polyethylene glycol; PLGA: poly(lactic-co-glycolic acid); PCL: polycaprolactone; DA: diacrylate; PLLA: poly(L-lactic acid); PEO: poly(2-ethyl-2-oxazoline).

Review

s� F

OUNDATION

REV

IEW

formulation where upon injection the gel undergoes phase trans-

formation by an external stimulus, such as temperature, pH, and

ionic composition. Once injected into the vitreous, the in situ

gelling system forms a depot where drug slowly diffuses/releases

from the gel formulation. The hydrogel can also be formed ex situ

into microparticles or in monolithic form and broken up for

injectability. The size, shape, stiffness and surface chemistry of

the hydrogel can be modified and engineered to increase biocom-

patibility. The injected hydrogel may cause clouding of the vitre-

ous if freely floating and non-transparent. If the density of the

hydrogel is slightly higher than that of the vitreous humor, it may

settle at the bottom of the vitreous and not disrupt the visual path.

However, physical contact with the retina could increase the risk

of retinal damage or other adverse findings, which must be

assessed through pre-clinical testing.

IVT injection of a novel hydrogel delivery system may hold

great promise for delivery of proteins to the back of the eye. Many

new hydrogel ocular delivery platforms have emerged in recent

years. A recent paper [40] showed in vivo sustained release of

bevacizumab from a thermal responsive gel made from poly(eth-

ylene glycol)-poly(serinol hexamethylene urethane) block copol-

ymer after IVT administration. The drug is physically encapsulated

by the polymer matrix and is released when the polymer backbone

degrades. Another study demonstrated an in situ-crosslinked poly-

saccharide for bevacizumab released by mixing oxidized alginate

and glycol chitosan [41]. Injection of the mixture into the vitreous

in the sol state resulted in gel formation and showed sustained

release of bevacizumab for up to 3 days. This relatively short release

duration may not be clinically viable, which demonstrates the

typical challenge for delivering biologic by physical entrapment

within the hydrogel matrix. Namely, can enough drug be loaded

into the delivery system and the release from the matrix be slowed

enough to enable long-term, i.e., multiple months, release of a

clinically relevant quantity, or effective dose of an ophthalmic

drug?

To prolong the release duration, some groups have explored

chemical conjugation of the drug to the hydrogel. Cleavable

linkers have been used to conjugate drugs to poly(ethylene glycol)

(PEG) hydrogels for half-life extension, where the drug release and

hydrogel degradation is tunable by varying the linker chemistry

[42]. A different linker design that is enzymatically responsive

showed release of therapeutic peptides from a PEG hydrogel upon

exposure to matrix metalloproteinase (MMP) [43]. Yu et al. showed

in-vivo 6-month release of bevacizumab from chemically cross-

linked hyaluronic hydrogel after intravitreal injection in rabbit

eye. The hydrogel showed good biocompatibility over the 6-

month time frame with no increase in intraocular pressure or

inflammation and retinal damage [44].

Suprachoroidal delivery from hydrogelsThe suprachoroidal space is an artificially created, or ‘‘potential”

space between the sclera and choroid, formed by injecting material

into the site, thereby separating these layers. Direct suprachoroidal

injection localizes the therapeutic agents at the site of action to

provide higher drug exposure to the choroid and retina. Injection

of a hydrogel sustained delivery system can limit the otherwise fast

clearance of biologics [45] through the choroid blood flow to

maintain constant drug exposure [46], while systemic exposure

to the administered drug is comparable with intravitreal injections

[36].

Due to the sieving effect, intact particles ranging from 20 nm to

10 mm that are injected into the suprachoroidal space remain

there for months [47]. This suggests that biodegradable hydrogel

particles can be used for sustained delivery by this route of admin-

istration, such that intact particles will stay in the space and clear

when degraded.

Studies with a light-activated in situ forming gel of poly(capro-

lactone dimethacrylate) (PCM) and hydroxyethyl methacrylate

(HEMA) showed sustained suprachoroidal delivery of bevacizu-

mab for up to 4 months in rat eyes [48]. However, the light-

activated system has the limitation of free radical generation that

could cause damage to the eye.

Subconjunctival delivery from hydrogelsLocalized drug delivery to the posterior segment of the eye can also

be achieved through subconjunctival injection of a hydrogel at the

space between the conjunctiva and sclera. Subconjunctival injec-

tions have greater bioavailability than topical administration [36]

and can be less invasive than intravitreal injection, and it has

combined merits of both administrations. Studies have demon-

strated sustained release of insulin with a subconjunctivally

implanted hydrogel [49]. These hydrogels are biodegradable and



Reviews�FO

UNDATIO

NREV

IEW

thermoresponsive, and showed sustained release for up to one

week, with no inflammation or evidence of toxicity in the in vivo

rat and in vitro retinal cell model.

To achieve efficacious delivery of a biologic into the retina from

a subconjunctival injection requires knowledge of protein pene-

tration across the intraocular tissue. Studies with C14 insulin

injection showed the dominant penetration pathway through

direct diffusion across the sclera into the posterior segment of

the eye [50].

To enhance drug penetration into the eye, transscleral ionto-

phoresis has been studied [51]. Iontophoresis uses small electric

currents to enhance charged molecule penetration. Antibiotic-

loaded hydrogels coupled with a short iontophoretic treatment

increased drug penetration to the eye [52]. However, delivery of

high MW compounds to the inner retina via transconjunctival

iontophoresis was not as effective in vivo [32].

Classes of injectable hydrogelsThe most effective way to maximize posterior protein bioavailabil-

ity is via injection into the vitreous body, the clear gel in the

vitreous chamber, or into other posterior spaces. As a result,

significant interest has been focused on designing dynamic hy-

drogel chemistries that can undergo a change in physicochemical

properties upon injection. While this is an extremely active area of

research in the broader hydrogel field, those that possess the

greatest promise for posterior ocular applications are stimulus-

responsive hydrogels, shear-thinning hydrogels, in situ gelling

systems and nano- or micro-gels.

Stimulus-responsive hydrogels are attractive for biomedical

applications because they can respond to specific chemical, physi-

cal or biological cues [53]. Hydrogels that undergo a physicochem-

ical change in response to pH, temperature or light take advantage

of the nonlinear response of certain chemical properties to these

stimuli. In this way, even relatively minor environmental changes

can result in a significant change in hydrogel properties.

Understanding how to take advantage of this stimulus-respon-

siveness is paramount in designing an effective system. In the

ocular realm, stimulus-responsiveness is generally used to achieve

a sol-gel transition between the liquid pre-injection state of the

system and the solid post-injection state. In this way, a solution

that is readily injected into the vitreous can then respond to, e.g.

the increase in temperature from room to body temperature and

solidify into a cohesive gel.

Thermosensitive hydrogelsTemperature-sensitive polymers are characterized by a critical

temperature at which their sol-gel behavior undergoes a drastic

change [28]. Polymers most useful for hydrogel applications are

those that exhibit a lower critical solution temperature (LCST);

these polymers behave as a solution below this temperature and

as a gel above it [54]. This transition is driven by the development

of hydrophobic chain entanglements, which result in phase

separation within the aqueous polymer solution as a physically

cross-linked network is formed. For biomedical applications,

ideal polymers are those that exhibit this transition between

room and body temperatures. In this way, the aqueous solution

can be injected at room temperature but rapidly transition to a gel

at body temperature.


Because this transition is defined by a shift between hydration

and hydrophobic interactions governing the polymer’s physical

state, there are some special considerations for applications to

drug delivery. First is that this transition can often result in the

expulsion of large amounts of water as the polymer phase

separates. For example, poly(N-isopropylacrylamide) (PNI-

PAAm)-based materials heated from 22 �C to 37 �C expel up

to 200 wt% of water [55]. This rapid deswelling can drive en-

capsulated hydrophilic molecules out of the matrix, resulting in

a large burst release, although this effect can be modulated by

incorporating hydrophilic molecules such as PEG into the poly-

mer matrix [56]. Temperature-sensitive polymers are water sol-

uble at lower temperatures because the hydrogen bonding of the

polymers with water is stronger than the hydrophobic interac-

tions between the polymers. As the temperature rises, however,

the hydrogen bonding becomes weaker, and the hydrophobic

interaction becomes stronger, resulting in precipitation of poly-

mers from solution or collapse of the hydrogel structure. This

temperature-dependent increase in hydrophobic character

within the hydrogel and/or subsequent network collapse could

pose a concern for immunogenicity [57] and antibody stability

[58,59], both of which should be carefully examined during

investigation of such systems.

In recent years, several ocular protein delivery systems based on

temperature-sensitive polymers have been reported. For example,

Derwent and Mieler reported in 2008 on a PEG-crosslinked PNI-

PAAm thermosensitive hydrogel for posterior ocular delivery of

proteins [60]. Rapid deswelling of the matrix at 37 �C resulted in a

large efflux of the encapsulated protein in the first few hours. In

addition, complete release of encapsulated proteins was not ob-

served, indicating some amount was entrapped within the poly-

mer matrix. In 2011, Wang and colleagues reported on a

thermosensitive hydrogel that could sustain the release of beva-

cizumab for approximately 10 days [61]. The material was well

tolerated in the vitreous and did not result in physiological

changes to the retinal tissues. In 2013, Park and colleagues

reported on a reverse thermal gelling polymer intended for intra-

vitreal delivery of bevacizumab [62]. The copolymer system, com-

posed of a tri-block PEGylated polyurethane, was injectable

through a 27G needle and transitioned from sol to gel between

32 and 39 �C. Release kinetics of bevacizumab in vitro showed a 10–

35% burst followed by sustained release over 17 weeks. In vivo

studies demonstrated an approximately 5-fold improvement in

intraocular bioavailability compared to bevacizumab alone, which

was sustained for 9 weeks [40].

A novel hybrid system developed by Patel and colleagues

entrapped protein-loaded nanoparticles within a thermosensitive

gelling polymer [63]. In vitro results demonstrated low cytotoxicity

and high biocompatibility in cell-based assays, while release of

model proteins could be sustained for 60 days with near-zero-order

kinetics. Entrapment of nanoparticles within the thermogel was

also found to reduce the burst release phenomenon of nanopar-

ticles alone, presumably due to the additional diffusion barrier

provided by the hydrogel. One challenge for these types of multi-

component systems is demonstration that sufficient drug can be

loaded within 50–100 mL of the formulation to provide a thera-

peutic level of drug release over the intended duration. In this

example, the authors achieved between 5 and 6% drug loading


Review

s� F

OUNDATION

REV

IEW

within the nanoparticle, and it is not clear what duration of

clinically meaningful drug exposure this would provide.

Photosensitive hydrogelsThat light can be transmitted directly into the posterior segment of

the eye in a controlled and targeted manner may open up possi-

bilities for the use of photosensitive or photocrosslinkable hydro-

gels. As in thermoresponsive polymers, hydrogel precursors could

presumably be injected into the vitreous as a solution. However,

instead of using temperature to affect gelation, a targeted laser or

other light source would be directed at the injected precursor to, e.

g. drive crosslinking via a photosensitive crosslinker.

Such a chemistry was demonstrated by Envisia, in which protein

microparticles were encapsulated within a photocurable hydrogel

[64]. In vitro, this system released bevacizumab for 90 days, while

its in vivo application and performance has yet to be reported.

Tyagi et al. also employed a light-activated hydrogel system for

posterior protein delivery using a suprachoroidal injection of

photocurable hydrogel precursors and free bevacizumab [48].

While this system could release the protein in vivo for several

months, the extended time needed for complete curing resulted

in a burst release of more than 20% of loaded protein.

While promising in principle, this approach does carry several

potential liabilities for an intravitreally-administered drug delivery

system. One such liability is the amount of time required to

crosslink the system upon injection and whether this can be done

quickly enough to prevent diffusion of precursor components

throughout the vitreous. A bigger concern may be the generation

of free radicals by photoinitiators. Produced in situ, these free

radicals may cause toxicity in nearby tissues, which would prohibit

their use. Further, the stability of drugs—especially protein thera-

peutics—would be a major concern in the presence of free radical-

generating photoinitiators, as well as after exposure to high inten-

sity UV irradiation.

In situ forming hydrogelsInstead of using biological stimuli to initiate gelation after injec-

tion, some groups have designed hydrogels that spontaneously

form after injection [65]. These in situ forming hydrogels form

through coupling of reactive species at the injection site. Since

there is no stimulus to activate this reaction, these systems are

typically formulated such that the various hydrogel precursors are

mixed immediately prior to the injection event. The injected pre-

gel then spontaneously reacts to form a cohesive network at the

site of injection.

In order to design a successful in situ gelling protein delivery

hydrogel, several key factors must be controlled. First, the rate of

reaction must be carefully tuned within a narrow window since

network formation must be: a) fast enough to react quickly upon

injection to encapsulate the protein and prevent escape of the

hydrogel precursors by diffusion away from the injection site; but

b) slow enough to allow injection of the pre-gel since viscosity of

this solution will increase with network formation and eventually

prevent injection.

In situ gelling protein delivery systems have several liabilities

that must be considered carefully during the design process [66].

First, diffusion of the protein drug out of the hydrogel before the

network is fully formed must be controlled in some way. Strategies

to address this will be explored in more detail in Section 5, but may

involve immobilization of the protein to the hydrogel precursors

in some manner. Second, in order for an in situ gelling system to be

injectable yet form a network upon injection, some reactive spe-

cies and/or catalysts must be introduced to the injection site (since

stimuli-responsive chemistries are excluded in this category). In

order for this to be feasible, reactive chemistries must be carefully

chosen that they don’t elicit an unwanted reaction in vivo. For

most intraocular tissues, this bar is relatively high and will exclude

most commonly used chemistries for consideration.

In general, the requirements of an in situ gelling hydrogel

crosslinking chemistry include: a) amenability to reaction in an

aqueous physiological environment; b) rapid but controllable

kinetics of reaction; c) non-cytotoxicity to surrounding or nearby

tissues; and d) non-destructivity to the protein drug being encap-

sulated within the gel. This relatively demanding list of require-

ments precludes most commonly used chemistries from serious

consideration for a protein-encapsulating hydrogel. However,

several more recent chemistries may be able to achieve this,

especially those that fall under the category of ‘‘bioorthogonal”

reactions [67]. These include the copper-free click reaction be-

tween azides and cyclooctynes and the tetrazine ligation with

norbornene or trans-cyclooctene.

Several injectable in situ gelling systems have been developed for

posterior protein delivery. In 2013, Xu et al. reported on an

alginate chitosan hydrogel for sustained release of bevacizumab

[41]. The release profile was characterized by a 20-30% burst release

in the first several hours associated with network formation and

complete release within 3 days. In 2014 Yu et al. reported on a PEG-

based hydrogel formed by the Michael addition reaction between

two tetrameric PEG molecules, PEG-maleimide and PEG-SH [68].

Pre-formed gels were found to release bevacizumab over several

weeks; however, the release profile of gels formed in situ was not

included.

Micro- and nanogelsThe aforementioned systems seek to overcome the limitation of

injectability by designing a system that can transition from a

solution to a gel during or after injection. Alternatively, hydrogels

can be made injectable by reducing their size to the nano- or

micro-scale, which would permit their direct injection into the

vitreous or other posterior sites. Hydrogels fabricated at this scale

are often referred to as nano- or microgels. This approach is being

explored for instance by Envisia Therapeutics, which uses their

proprietary PRINTTM technology to fabricate protein-loaded hy-

drogel microparticles [69].

While overcoming the challenge of injectability, this approach

does introduce a new challenge: the greatly reduced distance a

protein must diffuse to be released from the hydrogel. As a result,

novel release-modifying strategies become especially important to

sustain delivery on a usable time scale, as explored in Section 5.

Shear-thinning hydrogelsShear-thinning hydrogels are another class of injectable hydrogels

that takes advantage of transient or reversible interactions instead

of covalent bonds to provide crosslink points in the hydrogel

network. In the presence of high mechanical shear, such as within

a syringe needle during injection, these crosslinks are disrupted



Reviews�FO

UNDATIO

NREV

IEW

and the hydrogel can flow through the needle. In the absence of

shear forces, such as once the hydrogel is deposited within the

vitreous, the crosslink points can re-form, forming a cohesive

depot [70]. Unlike chemically crosslinked in situ gelling systems,

where precursors are mixed and covalent bonds are formed after

injection, the shear-thinning hydrogel can be formed ex vivo

because crosslinking is non-covalent. The non-covalent crosslink-

ing of the shear-thinning gel is typically through self-assembly

from physical associations such as hydrophobic interactions, hy-

drogen bonding, electrostatic interactions, and host-guest inter-

actions [71]. Examples of shear-thinning systems are beta-hairpin

peptide-based fibrillar hydrogel, recombinant multidomain pep-

tides and cyclodextrin-block copolymer mixtures [70]. A successful

shear-thinning system should self-assemble to form the intended

network structure at physiological conditions, flow upon injection

and self-heal after injection. Kinetics of network recovery after

injection and associated drug encapsulation efficiency need to be

tuned to ensure shear-thinning hydrogels can be a viable option

within a clinical setting.

Drug encapsulation and release strategiesMany sustained drug delivery systems control drug release by

limiting the ability of the molecule to diffuse out of the encapsu-

lating matrix. For example, poly(lactic-co-glycolic acid) (PLGA) is a

relatively dense hydrophobic matrix. The small mesh size charac-

teristic of matrices like PLGA results in small pores and high

tortuosity, which are able to limit free diffusion of molecules from

the matrix and thereby extend the timeframe over which the drug

is released [72]. While these systems have proven effective for

controlling the release of small molecule drugs, their hydrophobic

nature and relatively extreme manufacturing conditions (i.e. use

of organic solvents, high temperature, and/or high shear environ-

ment) make them more challenging to implement for delivery of

biologics [7].

While hydrogels are better suited to encapsulation of biologics,

most have mesh sizes orders of magnitude larger than hydropho-

bic polymers due in part to the swelling effects of water. As a result,

Diffusion

Degradation

Linker cleava

Chemical conj

Physical entra

(Time, pH, lig

Degradation

FIGURE 3

Drug encapsulation and release from hydrogel networks can be accomplished by

control strategies will be dictated by the therapeutic needs of the specific drug


they are not able to effectively limit diffusion of protein molecules

out of the matrix, resulting in relatively fast efflux of biologics out

of the system (time scales of hours to days are typical for standard

macroscopic hydrogel delivery systems). While this time scale may

be acceptable for some applications of protein delivery, many

require a much longer duration of release to meet the therapeutic

requirements.

In order to address this shortcoming, researchers have devised

numerous methods of controlling protein efflux to prolong the

time scale of release. These include slowing diffusion of the protein

through physical or chemical means, among others, and use of

permanent or transient linkages between the protein and the

hydrogel matrix, as illustrated in Fig. 3.

Diffusion-modulating strategiesThe driving force for rapid efflux of proteins from most hydrogels

is their largely unencumbered diffusion out of the hydrated hy-

drogel matrix. By extension, if the diffusion can be encumbered in

some way, that should prolong the time required for diffusion out

of the hydrogel thereby extending the duration of release. This

effect can be achieved by either physical or chemical means, as has

been demonstrated in several recent efforts.

The mechanism for physically limiting protein diffusion out of

a hydrogel is derived from limiting the effective diffusion coeffi-

cient of the molecule through the matrix. This effective diffusion

coefficient through a porous medium, De, is defined as:

De ¼ D et dt

where D is the diffusion coefficient of the protein in the medium

filling the pores (i.e. water), et is the porosity available for trans-

port, d is the constrictivity and t is the tortuosity. Within this

equation, porosity and constrictivity (and in practice tortuosity)

can all be used to decrease the effective diffusion coefficient by

decreasing the mesh size of the hydrogel (the mesh size of a

polymer matrix is defined as the average distance between neigh-

boring polymer strands).

ge

ugation

pment

ht)


physical entrapment or chemical conjugation. The decision between release being released.


Review

s� F

OUNDATION

REV

IEW

This strategy was employed recently by Tong et al. [73] to

produce a PEG-based hydrogel that could release a growth factor

for approximately two months using BSA as a model protein. In

this system, the mesh size of the hydrogel was decreased to be on

the order of the hydrodynamic radius of BSA (approximately 4

nm), which effectively immobilized the protein within the PEG

network. As the hydrogel network degraded hydrolytically over

time, the mesh size increased and the protein could slowly diffuse

out. This strategy permitted first-order drug release kinetics for

approximately 60 days using BSA as a model.

An important note in this approach is the requirement that the

protein be present during the hydrogel formation process since the

small mesh size won’t allow the protein to enter the network once

it is formed. As a result, the crosslinking or polymerization chem-

istry used must be non-degrading to the protein. Commonly used

photo-initiating species, such as those used by Tong et al., are likely

to chemically degrade proteins or attach them to the polymer

network [74]. These details are often overlooked in published

reports, but require careful attention from investigators to preserve

the biological availability and activity of released proteins.

An alternative strategy employed by Ocular Therapeutix has

been to encapsulate the protein within a relatively small particle,

which is then entrapped within a hydrogel matrix [75]. Since the

protein is not free to diffuse within the hydrogel, its release is

dictated by degradation of the hydrogel matrix and can be tuned as

such. This strategy—encapsulation of the protein within a matrix,

which is dispersed within a hydrogel—has also been explored by

other groups, e.g. using PLGA particles [76–78].

In addition to physical mechanisms of preventing free protein

diffusion through a hydrogel, electrostatic interactions have also

been used. In these hydrogels, charged groups are introduced such

that the hydrogel can electrostatically interact with the protein. If

these groups have charges opposite the charge on the protein at

physiological pH, the attractive forces resulting from this interac-

tion can slow diffusion of the protein out of the matrix. For

example, Purcell et al. incorporated sulfated hyaluronic acid

(HA) into HA gels to slow release of a highly positively charged

model protein [79]. Compared to the more neutral albumin,

whose release was not modified by incorporation of negative

sulfate groups, release of the positively charged protein was ap-

proximately 3-fold slower when sulfate groups were present.

From a practical standpoint, this approach is less broadly appli-

cable than other strategies since it relies entirely on the isoelectric

point of the protein of interest. If the target protein isn’t suffi-

ciently positively or negatively charged at physiological pH, ionic

interactions won’t be strong enough to provide a controlling effect

on their diffusion. In addition, the presence of salts in physiologi-

cal environments can screen this interaction, thereby reducing its

effect in vivo. For these reasons, this strategy may be challenging to

practically implement.

Chemical conjugationThe strategies described in the previous section to limit free

diffusion of proteins have been used in research fairly extensively,

but their translation to a clinical product has been slow due to the

numerous challenges described. In response to the challenges

presented by those approaches, researchers began exploring the

idea of chemically attaching the protein to the hydrogel backbone

in a reversible manner. The linker between the protein and the

hydrogel was made responsive to physiological stimulus (e.g. pH)

to release the protein at the desired rate largely independent of the

hydrogel properties.

This cleavable linker strategy is an area of active development

with several companies exploring this space. ProLynx, LLC. and

Ascendis Pharma are two examples of companies developing

linkers for drug delivery that can transiently attach a protein to

a hydrogel backbone and release the free drug over an extended

time under physiological conditions [42,80,81]. Critically, this

cleavable linker strategy liberates several hydrogel properties from

being dictated by protein release behavior. For example, since

diffusion of the protein isn’t a critical parameter in determining

its release behavior, hydrogels could be fashioned as nano- or

microparticles with minimal impact on the release kinetics. In

addition, mesh size, swelling and degradation properties can be

decoupled from release kinetics, further freeing up formulation

flexibility. For these reasons, this approach may have a significant

impact on the development and translation of future ocular

protein delivery hydrogels.

Development considerationsTo bring any ocular delivery technology to market, it is of para-

mount importance that the delivery vehicle is safe and that the

production is tightly controlled. Safety must be demonstrated in

pre-clinical animal testing before an investigational new drug

application (IND) is submitted and approved for clinical testing.

The production of the drug delivery system must be tightly con-

trolled so that it is consistent from batch to batch. These activities

are known as chemistry, manufacturing and control (CMC). The

following sections discuss some of the challenges that will need to

be considered for development of a successful ocular drug delivery

system.

Clinical considerationsOne of the most important requirements for any ocular delivery

technology is pre-clinical demonstration of safety and tolerability

of the entire delivery system, including the procedure of adminis-

tration. The retina in particular is a complex tissue in which minor

perturbations can adversely affect vision; therefore, a complete

lack of retinal toxicity is paramount. Severe and/or sustained

adverse reactions can lead to irreversible damage, e.g. retinal

detachment. Inflammation and foreign body reactions in all ocu-

lar tissues need to be quite limited or non-existent, as demonstrat-

ed by appropriately designed pre-clinical safety studies.

An important design aspect of any hydrogel intended for ocular

use is clearance of the gel from the site of injection, which should

ideally occur in a similar time frame to drug release. Since most

back of the eye diseases are chronic and would require continuous

re-administration of the drug depot, accumulation of hydrogel

components over time must be prevented. This can be achieved

through selection of a biodegradable polymer backbone, incorpo-

ration of degradable crosslinkers, or through engineered disassem-

bly of the hydrogel network over time (e.g. as is often the case for

shear-thinning hydrogel formulations). Relatedly, demonstration

of safety and tolerability of the injected hydrogel should be paired

with an assessment that the degradation products resulting from

hydrogel erosion or disassembly are also well tolerated.



Reviews�FO

UNDATIO

NREV

IEW

The intended form of the hydrogel formulation will also play a

critical role in defining associated pre-clinical testing require-

ments. As most hydrogels are cross-linked polymer networks, a

strategy for delivery must be established early in pre-clinical

testing. The injection force, modeled by the Poiseuille equation,

is inversely proportional to the fourth power of the needle radius

and directly proportional to the viscosity, injection speed and

square of syringe plunger radius [46]. To reduce the injection force,

the hydrogel solution viscosity must be low or be lowered during

injection, for example through use of a thermosensitive, shear-

thinning or in situ forming chemistry. In all of these cases, pre-

clinical testing should be used to confirm reproducible recovery of

hydrogel properties after injection. Alternatively, hydrogels may

be constructed as nano- or micro-gels to facilitate injection into

the eye. Based on previous reports, careful attention must be given

to the fate of injected particles as there is evidence that such

particles are mobile and can migrate outside of the intended tissue

[82]. For in situ forming hydrogels that crosslink upon injection,

the mixing of the gel precursors and the time to cohesive gelation

are two important factors that need to be evaluated before ad-

vancement to clinical testing.

Chemistry, Manufacturing and ControlsAll aspects of the chemistry and manufacturing process of the drug

delivery system need to be well defined with appropriate, validated

analytical methods to monitor the product. The International

Conference on Harmonization (ICH) has set guidelines on the

specifications, test procedures and acceptance criteria for new drug

substances and drug products for small molecules and biologics.

Specifications are part of a control strategy to ensure product

quality and consistency. For a novel drug delivery system, new

test procedures need to be established and validated. For hydrogels

derived from natural polymers such as alginate or hyaluronic acid,

raw material specifications and thorough characterization of im-

purities will be critically important to ensure proper performance

and control.

Manufacturability of the hydrogel delivery system is another

important challenge. Any laboratory scale synthesis will need to be

scaled up and all the handling and processing will need to be

evaluated. If the drug needs to be chemically conjugated to the

hydrogel, then the conjugation process and scalability will need to

be addressed. The protein therapeutic needs to be stable during the

conjugation process and at the same time any chemicals used

during the synthesis step will need to be removed, as any residual

materials may affect tolerability in the eye.

For all ophthalmic delivery systems, the United States Pharma-

copeia (USP) has set guidelines that must be followed. The mono-

graph Ophthalmic Preparations – Quality Tests <771> describes

the quality tests and that should be applied [83]. Hydrogels are

described as novel ophthalmic dosage forms, which must follow

the general ophthalmic delivery guidelines. Ophthalmic drug

delivery products must meet specifications for all general quality

attributes such as identity, purity, potency, sterility and particulate

matter. In addition, any drug-releasing system will need controls

and assays in place to evaluate lot-to-lot drug release performance

and variability. Establishment of accelerated release and/or


degradation assays early in the development process may greatly

facilitate development activities.

Sterility, bioburden and pyrogenic impurity testing are critical

aspects of drug product quality testing. For hydrogel delivery

systems, any sterilization process must not induce degradation

or morphological changes to the hydrogel or drug components.

Depending on the hydrogel system, autoclaving at elevated tem-

perature may be an option, but protein stability will likely be

adversely affected through this harsh process. Sterile filtration is

only applicable to formulations with particulates less than 0.2 mm

in diameter. Many cross-linked hydrogels with conjugated or

encapsulated biologics will require an aseptic manufacturing pro-

cess after sterile filtration of liquid components and sterilization of

other process inputs to achieve a sterile drug product.

Information on all aspects of the CMC process will be required

for submission to FDA to ensure proper identity, quality, strength/

potency, and purity of the drug substance and drug product.

Physicochemical characterization of the hydrogel drug product

such as morphology, size, viscosity and in vitro/in vivo degradation

will be critically useful for regulatory filings. Stability studies of the

hydrogel drug product will be needed to establish storage condi-

tions and shelf life. Development of any ocular drug delivery

system will be a technical and clinical challenge, but it could have

a significant impact in improving treatment for many patients.

ConclusionsOver the past decades, there have been considerable developments

and advances in ocular macromolecular therapeutics. Anti-VEGF

therapies for wet AMD set the stage for development of more

macromolecule therapeutics treating back of the eye diseases.

Many such therapeutics are currently in development, increasing

the need for innovative ocular delivery methods. Sustained deliv-

ery of biologics with hydrogels has a promising future because of

the highly biocompatible and protein-compatible nature of many

hydrogel chemistries. The physiochemical properties of the hy-

drogel can be engineered to respond to external stimuli such as pH,

temperature and light. These stimulus-responsive properties can

be used to make hydrogels injectable and allow for localized

delivery to the back of the eye. Drug encapsulation and release

strategies for hydrogel delivery systems include physical entrap-

ment and chemical conjugation. The major challenge now is to

design a hydrogel system that can provide sustained drug release

for several months and is completely physically cleared at the end

of drug release, or relatively soon thereafter, to avoid matrix

accumulation. The future for hydrogel drug delivery systems is

promising, but advances must be made through interdisciplinary

collaborations to address some of the important challenges such as

clinical translation, safety and tolerability, drug stability, and

manufacturability. Overcoming these remaining hurdles will al-

low hydrogel ocular drug delivery systems to realize their full

potential to treat otherwise debilitating ocular diseases while

reducing the burden of treatment for patients and clinicians.

Conflicts of InterestThe authors declare that there are no competing financial inter-

ests.


References

Review

s� F

OUNDATION

REV

IEW

1 Eljarrat-Binstock, E. et al. (2010) New Techniques for Drug Delivery to the Posterior

Eye Segment. Pharm. Res. 27, 530–543

2 Joseph, M. et al. (2016) Recent perspectives on the delivery of biologics to back of the

eye. Expert Opin. Drug Deliv. 14, 631–645

3 Tabandeh, H. et al. (2007) Intraocular Pressure Following Intravitreal Injection.

Invest. Ophthalmol. Vis. Sci. 48, 3374

4 Hutton-Smith, L.A. et al. (2017) Ocular Pharmacokinetics of Therapeutic Antibodies

Given by Intravitreal Injection: Estimation of Retinal Permeabilities Using a 3-

Compartment Semi-Mechanistic Model. Mol. Pharm. 14, 2690–2696

5 Frokjaer, S. and Otzen, D.E. (2005) Protein drug stability: a formulation challenge.

Nat. Rev. Drug Discov. 4, 298–306

6 Lee, J.-H. and Kim, H.-W. (2018) Emerging properties of hydrogels in tissue

engineering. J. Tissue Eng. 9, 204173141876828

7 Houchin, M.L. and Topp, E.M. (2008) Chemical Degradation of Peptides and

Proteins in PLGA: A Review of Reactions and Mechanisms. J. Pharm. Sci. 97,

2395–2404

8 Sharpe, L.A. et al. (2014) Therapeutic applications of hydrogels in oral drug delivery.

Expert Opin. Drug Deliv. 11, 901–915

9 Peppas, N. (2000) Hydrogels in pharmaceutical formulations. Eur. J. Pharm.

Biopharm. 50, 27–46

10 Zhu, J. and Marchant, R.E. (2014) Design properties of hydrogel tissue-engineering

scaffolds. Expert Rev. Med. Devices 8, 607–626

11 Vermonden, T. et al. (2012) Hydrogels for Protein Delivery. Chem. Rev. 112,

2853–2888

12 Li, J. and Mooney, D.J. (2016) Designing hydrogels for controlled drug delivery. Nat.

Rev. Mater. 1, 12 16071

13 Bae, K.H. and Kurisawa, M. (2016) Emerging hydrogel designs for controlled protein

delivery. Biomater. Sci. 4, 1184–1192

14 Delplace, V. et al. (2019) Controlled release strategy designed for intravitreal protein

delivery to the retina. J. Control Release 293, 10–20

15 Iyer, S. et al. (2019) Long-acting intraocular Delivery strategies for biological therapy

of age-related macular degeneration. J. Control Release 296, 140–149

16 Famili, A. and Rajagopal, K. (2017) Bio-Orthogonal Cross-Linking Chemistry

Enables In Situ Protein Encapsulation and Provides Sustained Release from

Hyaluronic Acid Based Hydrogels. Mol. Pharm. 14, 1961–1968

17 Ferreira, N.N. et al. (2018) Recent advances in smart hydrogels for biomedical

applications: From self-assembly to functional approaches. Eur. Polym. J. 99,

117–133

18 Flory, P.J. (1953) Principles of Polymer Chemistry. Cornell University Press, Ithaca, NY

19 Park, H. et al. (1993) Biodegradable hydrogels for drug delivery. CRC Press, Boca Raton,

FL

20 Schneider, E.A. (1894) Uber einige Methoden zur Abscheidung der Phosphorsaure

als phosphorsaures Alkali aus den Phosphaten des Kalkes und Eisenoxydes.

Zeitschrift Fur Anorganische Chemie 7, 386–391

21 van Bemmelen, J.M. (1897) Die Absorption. Das Wasser in den Kolloıden, besonders

in dem Gel der Kieselsaure. Zeitschrift Fur Anorganische Chemie 13, 233–356

22 On the mechanism of gelation in reversible colloidal systems. Proc. R. Soc. London

66, 95–109

23 Wichterle, O. and LIM, D. (1960) Hydrophilic Gels for Biological Use. Nature 185,

117–118

24 Qiu, Y. and Park, K. (2001) Environment-sensitive hydrogels for drug delivery. Adv.

Drug Deliv. Rev. 53, 321–339

25 Buwalda, S.J. et al. (2014) Hydrogels in a historical perspective: from simple

networks to smart materials. J. Control Release 190, 254–273

26 Klouda, L. and Mikos, A.G. (2008) Thermoresponsive hydrogels in biomedical

applications. Eur. J. Pharm. Biopharm. 68, 34–45

27 Thrimawithana, T.R. et al. (2012) Environment-sensitive polymers for ophthalmic

drug delivery. J. Drug Deliv. Sci. Technol. 22, 117–124

28 Jeong, B. et al. (2012) Thermosensitive sol–gel reversible hydrogels. Adv. Drug Deliv.

Rev. 64, 154–162

29 Kim, Y.C. et al. (2014) Ocular delivery of macromolecules. J. Control Release 190,

172–181

30 Cholkar, K. et al. (2013) Eye: anatomy, physiology and barriers to drug delivery. In

Ocular Transporters and Receptors (Mitra, A.K., ed.), pp. 1–36, Woodhead Publishing,

Cambridge, UK

31 Subrizi, A. et al. (2019) Design principles of ocular drug delivery systems: importance

of drug payload, release rate, and material properties. Drug Discov. Today (in press)

32 Sanharawi El, M. et al. (2010) Protein delivery for retinal diseases: From basic

considerations to clinical applications. Prog. Retinal Eye Res. 29, 443–465

33 Ottiger, M. et al. (2009) Efficient Intraocular Penetration of Topical Anti–TNF-aSingle-Chain Antibody (ESBA105) to Anterior and Posterior Segment without

Penetration Enhancer. Invest. Ophthalmol. Vis. Sci. 50, 779–786

34 Sanharawi El, M. et al. (2010) Protein delivery for retinal diseases: from basic

considerations to clinical applications. Prog. Retinal Eye Res. 29, 443–465

35 Lim, M. et al. (2009) The Boston Ocular Surface Prosthesis as a Novel Drug Delivery

System for Bevacizumab. Semin. Ophthalmol. 24, 149–155

36 Nomoto, H. et al. (2009) Pharmacokinetics of Bevacizumab after Topical,

Subconjunctival, and Intravitreal Administration in Rabbits. Invest. Ophthalmol. Vis.

Sci. 50, 4807–4813

37 Schultz, C. et al. (2011) Drug delivery to the posterior segment of the eye through

hydrogel contact lenses. Clin. Exp. Optometry 94, 212–218

38 Zhang, L. et al. (2013) Zwitterionic hydrogels implanted in mice resist the foreign-

body reaction. Nat. Biotechnol. 31, 553–556

39 Jager, R.D. et al. (2004) Risks of intravitreous injection: a comprehensive review.

Retina 24, 676–698

40 Rauck, B.M. et al. (2014) Biocompatible Reverse Thermal Gel Sustains the Release of

Intravitreal Bevacizumab In Vivo. Invest. Ophthalmol. Vis. Sci. 55, 469–476

41 Xu, X. et al. (2013) Sustained release of avastin1 from polysaccharides cross-linked

hydrogels for ocular drug delivery. Int. J. Biol. Macromol. 60, 272–276

42 Ashley, G.W. et al. (2013) Hydrogel drug delivery system with predictable and

tunable drug release and degradation rates. Proc. Natl. Acad. Sci. U. S. A. 110,

2318–2323

43 Van Hove, A.H. et al. (2014) Development and in vitro assessment of enzymatically-

responsive poly(ethylene glycol) hydrogels for the delivery of therapeutic peptides.

Biomaterials 35, 9719–9730

44 Yu, Y. et al. (2015) Injectable Chemically Crosslinked Hydrogel for the Controlled

Release of Bevacizumab in Vitreous: A 6-Month In Vivo Study. Trans. Vis. Sci. Tech.

4, 5

45 Mansoor, S. et al. (2012) Pharmacokinetics and Biodistribution of Bevacizumab

Following Suprachoroidal Injection into the Rabbit Eye Using a Microneedle. Invest.

Ophthalmol. Vis. Sci. 53, 498

46 Einmahl, S. et al. (2002) Evaluation of a Novel Biomaterial in the Suprachoroidal

Space of the Rabbit Eye. Invest. Ophthalmol. Vis. Sci. 43, 1533–1539

47 Patel, S.R. et al. (2012) Targeted Administration into the Suprachoroidal Space Using

a Microneedle for Drug Delivery to the Posterior Segment of the Eye. Invest

Ophthalmol. Vis. Sci. 53, 4433–4441

48 Tyagi, P. et al. (2013) Light-Activated, In Situ Forming Gel for Sustained

Suprachoroidal Delivery of Bevacizumab. Mol. Pharm. 10, 2858–2867

49 Misra, G.P. et al. (2009) Subconjunctivally implantable hydrogels with degradable

and thermoresponsive properties for sustained release of insulin to the retina.

Biomaterials 30, 6541–6547

50 Lee, T.W.-Y. and Robinson, J.R. (2001) Drug Delivery to the Posterior Segment of the

Eye: Some Insights on the Penetration Pathways after Subconjunctival Injection. J.

Ocul. Pharmacol. Ther. 17, 565–572

51 Eljarrat-Binstock, E. et al. (2010) New Techniques for Drug Delivery to the Posterior

Eye Segment. Pharm. Res. 27, 530–543

52 Eljarrat-Binstock, E. et al. (2004) Delivery of Gentamicin to the Rabbit Eye by Drug-

Loaded Hydrogel Iontophoresis. Invest. Ophthalmol. Vis. Sci. 45, 2543–2548

53 las Heras Alarcon de, C. et al. (2005) Stimuli responsive polymers for biomedical

applications. Chem. Soc. Rev. 34, 276–285

54 Famili, A. et al. (2014) A Combined Micelle and Poly(Serinol Hexamethylene Urea)-

Co-Poly(N-Isopropylacrylamide) Reverse Thermal Gel as an Injectable Ocular Drug

Delivery System. Macromol. Biosci. 14, 1719–1729

55 Forney, B.S. and Guymon, C.A. (2011) Fast Deswelling Kinetics of Nanostructured

Poly(N-Isopropylacrylamide) Photopolymerized in a Lyotropic Liquid Crystal

Template. Macromol. Rapid Commun. 32, 765–769

56 Drapala, P.W. et al. (2014) The Effect of Glutathione as Chain Transfer Agent in

PNIPAAm-Based Thermo-responsive Hydrogels for Controlled Release of Proteins.

Pharm. Res. 31, 742–753

57 Moyano, D.F. et al. (2012) Nanoparticle Hydrophobicity Dictates Immune

Response. J. Am. Chem. Soc. 134, 3965–3967

58 Nakanishi, K. et al. (2001) On the adsorption of proteins on solid surfaces, a

common but very complicated phenomenon. J. Biosci. Bioeng. 91, 233–244

59 Kleijn, J.M. and Norde, W. (1995) The adsorption of proteins from aqueous solution

on solid surfaces. Heterog. Chem. Rev. 2, 157–172

60 Kang Derwent, J.J. and Mieler, W.F. (2008) Thermoresponsive hydrogels as a new

ocular drug delivery platform to the posterior segment of the eye. Trans. Am.

Ophthalmol. Soc. 106, 206–213


http://refhub.elsevier.com/S1359-6446(18)30400-8/sbref0005











































































































































Reviews�FO

UNDATIO

NREV

IEW

61 Wang, C.-H. et al. (2012) Extended Release of Bevacizumab by Thermosensitive

Biodegradable and Biocompatible Hydrogel. Biomacromolecules 13, 40–48

62 Park, D. et al. (2013) An Anti-angiogenic Reverse Thermal Gel as a Drug-Delivery

System for Age-Related Wet Macular Degeneration. Macromol. Biosci. 13,

464–469

63 Patel, S.P. et al. (2014) Tailor-Made Pentablock Copolymer Based Formulation for

Sustained Ocular Delivery of Protein Therapeutics. J. Drug Deliv. 1–15

64 Williams, S. et al. (2014) Photocurable Hydrogel Implants for the Extended Release

of Bevacizumab for the Treatment of Age Related Macular Degeneration. Invest.

Ophthalmol. Vis. Sci. 55, 471

65 Madan, M. et al. (2009) In Situ Forming Polymeric Drug Delivery Systems. Indian J.

Pharm. Sci. 71, 242–251

66 Van Tomme, S.R. et al. (2008) In situ gelling hydrogels for pharmaceutical and

biomedical applications. Int. J. Pharm. 355, 1–18

67 Patterson, D.M. et al. (2014) Finding the Right (Bioorthogonal) Chemistry. ACS

Chem. Biol. 9, 592–605

68 Yu, J. et al. (2014) In situ covalently cross-linked PEG hydrogel for ocular drug

delivery applications. Int. J. Pharm. 470, 151–157

69 Herlihy, K.P. et al. (2014) Extended Release of Microfabricated Protein Particles from

Biodegradable Hydrogel Implants for the Treatment of Age Related Macular

Degeneration. Invest. Ophthalmol. Vis. Sci. 55, 1960

70 Guvendiren, M. et al. (2012) Shear-thinning hydrogels for biomedical applications.

Soft Matter 8, 260–272

71 Jalalvandi, E. and Shavandi, A. (2019) Shear thinning/self-healing hydrogel based

on natural polymers with secondary photocrosslinking for biomedical applications.

J. Mech. Behav. Biomed. Mater. 90, 191–201

72 Sandor, M. et al. (2001) Effect of protein molecular weight on release from micron-

sized PLGA microspheres. J. Control Release 76, 297–311

73 Tong, X. et al. (2015) Long-Term Controlled Protein Release from Poly(Ethylene

Glycol) Hydrogels by Modulating Mesh Size and Degradation. Macromol. Biosci. 15,

1679–1686

74 Lin, C.-C. et al. (2007) Free-Radical-Mediated Protein Inactivation and Recovery

during Protein Photoencapsulation. Biomacromolecules 9, 75–83

75 Elhayek, R.F. et al. (2015) Pharmacokinetics of Bevacizumab Sustained Release from

Intravitreal Hydrogel Depots in a Rabbit Model Compared to a single Avastin Dose.

Invest. Ophthalmol. Vis. Sci. 56, 1522


76 Osswald, C.R. and Kang-Mieler, J.J. (2015) Controlled and Extended Release of a

Model Protein from a Microsphere-Hydrogel Drug Delivery System. Ann. Biomed.

Eng. 43, 2609–2617

77 Liu, W. et al. (2018) Biodegradable Microsphere-Hydrogel Ocular Drug Delivery

System for Controlled and Extended Release of Bioactive Aflibercept In Vitro. Curr.

Eye Res. 44, 264–274

78 Liu, W. et al. (2019) Characterization of Biodegradable Microsphere-Hydrogel

Ocular Drug Delivery System for Controlled and Extended Release of Ranibizumab.

Trans. Vis. Sci. Tech. 8, 12

79 Purcell, B.P. et al. (2014) Incorporation of sulfated hyaluronic acid macromers into

degradablehydrogel scaffolds forsustainedmoleculedelivery. Biomater.Sci.2,693–702

80 Santi, D.V. et al. (2012) Predictable and tunable half-life extension of therapeutic

agents by controlled chemical release from macromolecular conjugates. Proc. Natl.

Acad. Sci. U. S. A. 109, 6211–6216

81 Hersel, U. et al. AS AP. Hydrogel formulations. Ascendis Pharma AS. US9056126B2.

82 Adamson, P. et al. (2016) Single ocular injection of a sustained-release anti-VEGF

delivers 6 months pharmacokinetics and efficacy in a primate laser CNV model. J.

Control Release 244, 1–13

83 Ophthalmic Products-Quality Tests (2018) United States Pharmacopeia and National

Formulary USP -NF. MD, Rockville

84 Liu, W. et al. (2008) Alginate microsphere-collagen composite hydrogel for ocular

drug delivery and implantation. J. Mater. Sci. Mater. Med. 19, 3365–3371

85 Zhang, J. et al. (2013) A nanocomposite contact lens for the delivery of hydrophilic

protein drugs. J. Mater. Chem. B 1, 4388–4395

86 Huang, G. et al. (2004) Controlled drug release from hydrogel nanoparticle

networks. J. Control Release 94, 303–311

87 Baldwin, A.D. and Kiick, K.L. (2013) Reversible maleimide-thiol adducts yield

glutathione-sensitive poly(ethylene glycol)-heparin hydrogels. Polym. Chem. 4,

133–143

88 Agrahari, V. et al. (2016) Composite Nanoformulation Therapeutics for Long-Term

Ocular Delivery of Macromolecules. Mol. Pharm. 13, 2912–2922

89 Osswald, C.R. and Kang-Mieler, J.J. (2016) Controlled and Extended In Vitro Release

of Bioactive Anti-Vascular Endothelial Growth Factors from a Microsphere-

Hydrogel Drug Delivery System. Curr. Eye Res. 41, 1216–1222

90 Osswald, C.R. et al. (2017) In Vivo Efficacy of an Injectable Microsphere-Hydrogel

Ocular Drug Delivery System. Curr. Eye Res. 42, 1293–1301








































































Documents

Hydrogels for sustained delivery of biologics to the …kinampark.com/KPTopics/files/Hydrogels, General/Chang...REVIEWS Drug Discovery Today Volume 24,Number 8 August 2019 Teaser significant